2016
DOI: 10.1093/bioinformatics/btw350
|View full text |Cite
|
Sign up to set email alerts
|

A simple model predicts UGT-mediated metabolism

Abstract: The UGT metabolism predictor developed in this study is available at http://swami.wustl.edu/xenosite/p/ugt CONTACT: : swamidass@wustl.eduSupplementary information: Supplementary data are available at Bioinformatics online.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
43
0

Year Published

2017
2017
2024
2024

Publication Types

Select...
5
1

Relationship

0
6

Authors

Journals

citations
Cited by 57 publications
(43 citation statements)
references
References 22 publications
0
43
0
Order By: Relevance
“…In contrast to its predecessor, outputs are represented as probabilities rather than rank orderings. Xenosite is available as a web server (Matlock, Hughes, & Swamidass, 2015) and was extended to non-CYP applications such as UGT-mediated metabolism prediction (Dang, Hughes, Krishnamurthy, & Swamidass, 2016).…”
Section: Machine Learningmentioning
confidence: 99%
“…In contrast to its predecessor, outputs are represented as probabilities rather than rank orderings. Xenosite is available as a web server (Matlock, Hughes, & Swamidass, 2015) and was extended to non-CYP applications such as UGT-mediated metabolism prediction (Dang, Hughes, Krishnamurthy, & Swamidass, 2016).…”
Section: Machine Learningmentioning
confidence: 99%
“…The contemporary prediction methods have a good accuracy for reactions catalyzed by certain enzymes or enzyme classes, like the cytochrome P450 (CYP) superfamily (for examples, see references ). Recently, advances in non‐CYP SoM prediction have been made, and models for selected phase II reactions (glutathionation, glucuronidation) have been published . However, we are aware of only two holistic SoM models that can predict both phase I and phase II metabolism and are able to generalize over different enzyme families .…”
Section: Introductionmentioning
confidence: 99%
“…R ecently,a dvances in non-CYP SoM prediction have been made, [10] and modelsfor selected phase II reactions (glutathionation, glucuronidation) have been published. [11,12] However,w ea re aware of only two holistic SoM models that can predict both phase Ia nd phase II metabolism and are able to generalize over differente nzyme families. [13,14] Prediction tools that have ab road scope often rely on ligandbased machine learning strategies.…”
Section: Introductionmentioning
confidence: 99%
“…This work was based off an earlier model, Xenosite, an ANN‐based model for P450 metabolism on small‐molecules, which despite being a shallow network, was already outperforming the accuracy of SVM‐based models by as much as 5% . Further improvements were subsequently achieved by investigating the effect of using different types of molecular fingerprints for modeling P450 metabolism, where they discovered that further accuracy gains can be achieved using a consensus model utilizing different fingerprint types, and a related sister model that predicted the site of glucoronidation metabolism . In their more recent work on predicting epoxide‐based toxicity, Swamidass and coworkers designed a 4‐layer DNN architecture, and trained the model on a database of 702 epoxidation reactions, and identified SOEs with 94.9% AUC performance, and separated (i.e., classified) epoxidized and non‐epoxidized molecules with 79.3% AUC .…”
Section: Computer‐aided Drug Designmentioning
confidence: 99%